Correct usage of ocean tide data and height predictions from tide models requires understanding the coordinate system in which the ocean surface is measured, and the different contributions to tide height. Depending on the measurement technique, and corrections already applied to a data set of heights, tide height will consist of one or more of the following components:

- Ocean tide
- Earth body tide
- Ocean tide loading
- Self-attraction

The principal data set provided here describes the __ocean tide__, which is the perturbation of the ocean free surface elevation relative to the seabed. This is the usual product from ocean tide models, thus a direct comparison between the ocean tide database and ocean tide models is appropriate. Measurements made with bottom pressure recorders (BPRs) and coastal tide gauges (CTGs) are recorded with a seabed datum, with an offset to provide a zero mean tide.

Many elevation measurements, however, are obtained using geocentric coordinates, which are relative to the Earth’s center of mass. These measurements (GPS and satellite altimetry) must be corrected for Earth body tides (due to the direct attraction of the Sun and Moon on the deformable solid Earth) and __ocean tide loading (OTL) and self-attraction displacements__. Earth body tides are well known and accurately modeled in GPS and altimetry software and hence do not require further correction. OTL is the deflection of the deformable seabed by the tide-induced anomalous weight of water above it; see, e.g., Baker (1984). Self-attraction is the adjustment to the gravitational force caused by the redistribution of ocean mass in the vicinity of the point for which a prediction is required.

Self-attraction and OTL components are often combined into a “self-attraction and loading” (SAL) term. A very coarse approximation for SAL is that it is about 5% of the amplitude of, and 180^{o} out of phase with, the local ocean tide, i.e., the geodetic amplitude for each tidal constituent is about 95% of the ocean tide amplitude. Imagine the weight of water due to a 1 m tide (say) above mean sea level depressing the seabed by 0.05 m, so that the geodetic measurement of the tide is 0.95 m. However, the OTL displacement has larger length scales than the ocean tide, and so, especially close to the coast, it can vary significantly from this simplification. Using ocean tide models, Green’s functions and an Earth model, OTL displacements may be computed directly, as outlined below.

Gravimetric measurements measure gravity changes induced by the ocean tide plus those induced by the Earth body tides and OTL. Again, Earth body gravity tides are well known. The gravimetric records described here were originally converted from units of gravity (gals) to units of length using free-air and Bouger corrections; see, for example, *Williams and Robinson* (1980) or delta_h(m) = -3.768 delta_g(mgal). To be self-consistent, the OTL corrections outlined below follow the same approach, with modelled gravity OTL values obtained and then converted to length units. Once corrected for OTL, the major remaining error in the gravity record will relate to the calibration of the gravity meters, possibly inducing biases as large as 3-4% (pers. comm. T. Baker, 2005) of the total observed tide (prior to correction for Earth body tides and OTL).

In this database we only provide ocean tide coefficients. The GPS and gravity records have been corrected for SAL; the tiltmeter records have not. For SAL, we recommend the TPXO7.2 ocean load tide model for Antarctic OTL calculations. Thus, we use TPXO7.2 to generate SAL tidal coefficients for all sites for which we have ocean tide data using the SPOTL software (Agnew, 1997) . In some cases the SAL values are the same as those used to convert observed tide (e.g., for GPS) to ocean tide. For BPR and CTG data, which are obtained from direct analyses of sea surface elevation, the SAL model provides a mechanism for converting ocean tide to the elevations that should be measured by satellite altimeters.

**References**

Agnew, D.C., 1997. NLOADF: A program for computing ocean-tide loading, *J. Geophys. Res.*, **102** (B3), 5109-5110.

Baker, T. F., 1984. Tidal deformations of the Earth, *Science Progress*, **69**, 197-233.

Williams, R. T., and E. S. Robinson,1980. The ocean tide in the southern Ross Sea, *J. Geophys. Res.*, **85** (11), 6689-6696.